Study Design
The cohort was detailed in our previous research(16), and was designed to detect the association between parenteral anticoagulation therapy and clinical outcomes in patients with NSTE-ACS undergoing PCI. In brief, 8197 patients with NSTE-ACS undergoing PCI were enrolled from 5 centres from January 1,2010 to December 31,2014. Patients aged 18 years or older who were diagnosed with MCAD, and whose NT-proBNP level was determined on the first day of admission were included, cardiac arrest with return of circulation were included also. MCAD was defined as lesions with ≥ 50% diameter stenosis in the LM artery or ≥ 2 major coronary vessels with ≥ 50% stenosis. The exclusion criteria were as follows: pregnancy and missing baseline NT-proBNP. They were divided into 4 groups depending on the quartiles of NT-proBNP. Ultrasonic cardiography was conducted after admission, using Simpson's biplane method to calculate the left ventricular ejection fraction (LVEF). The estimated glomerular filtration (eGFR) rate was calculated using the Modification of Diet in Renal Disease equation based on Chinese patients(17). The study protocol was approved by the central ethics committee of Guangdong Provincial People’s Hospital, Guangzhou, China. The study was conducted in accordance with the Declaration of Helsinki.
Data Collection
The data were obtained in the first interview when the patient was admitted to the hospital. Baseline characteristic data were recorded by the responsible nurse or doctor; the baseline characteristics included demographic data and medical history. The procedural information originated from the catheterization report. All laboratory examinations were conducted during the first 24 hours after admission and before the procedure for all patients, and NT-proBNP was measured using an electrochemiluminescence immunoassay (Roche Diagnostics, Germany). All patients received the drug eluting stent. All interventional strategies were performed at the discretion of the heart team. In-hospital and follow-up assessments were performed by clinic visits or telephone interviews from November 7,2015 through December 30, 2016.
Outcomes
The primary outcome was in-hospital all-cause death. The secondary outcomes were all-cause death during the 3-year follow-up as well as in-hospital major adverse cardiovascular events (MACE), defined as a composite of all-cause death, myocardial infarction and stroke. The definitions of all clinical complications assessed during follow-up were identical to the original registry(16). Death was defined as all-cause deaths regardless cardiac or non-cardiac according to death records. Myocardial infarction was defined as classical symptoms accompanied by elevation of cardiac injury biomarker according to the third Universal Definition of Myocardial Infarction. Any stroke is defined as the presence of a new focal neurologic deficit thought to be vascular in origin, with signs or symptoms lasting more than 24 hours. A clinical events committee evaluated all clinical outcomes independently.
Statistical Analysis
Statistical analysis was performed using SAS version 9.4 (SAS Institute, Cary, NC, USA). Continuous variables are presented as the mean ± standard deviation. Categorical variables are presented as absolute and relative frequencies. Continuous variables were compared between groups using Student’s t-test (parametric variables). Multivariate regression analyses were carried out to evaluate the predictive value of NT-proBNP for different clinical outcomes, which was included as a continuous variable after logarithmic transformation. All the confounders included in the final model were either significant in the univariate analyses or clinical important factors. We included log NT-proBNP, Anaemia, Chronic heart failure, Chronic kidney disease, NSTEMI, LVEF and age to the final model of death analysis, and included log NT-proBNP, Anaemia, Chronic heart failure, chronic kidney disease, NSTEMI, LVEF, age, Diabetes, Myocardial infarction and operation time to the final model of MACE analysis. Receiver-operating characteristic (ROC) curves were used to assess the ability that NT-proBNP discriminates between patients who died and patients who survived during hospitalization. We also used the Youden index to determine the best cutoff of NT-proBNP for predicting all-cause death, and we expected to use this level for further analyses. Cumulative event analyses were performed to compare the long-term prognosis between patients who were divided by the best cutoff level of NT-proBNP. All P values <0.05 were considered statistically significant.